Unexplained High Fever in an Elderly Patient Treated With Clonidine, Duloxetine, and Atorvastatin

Abstract

Background:

Drug-induced fever is a clinical diagnosis and should always be considered when the fever is constant and high without a clear source of infection. Although drug-induced fever has been reported with other centrally acting antihypertensive drugs such as methyldopa, published reports of this adverse effect with clonidine in humans were not identified in a search of the literature.

Case summary:

A 66-year-old institutionalized white female with a history of morbid obesity (body mass index, 40 kg/m²), Alzheimer’s dementia, hypertension, and depression presented to a hospital in Boston, Massachusetts (Caritas Saint Elizabeth’s Medical Center) with generalized weakness and shortness of breath and was found to have a non–ST segment elevation myocardial infarction. Before hospitalization, the patient was taking memantine 10 mg PO BID, donepezil 10 mg PO once daily, duloxetine 60 mg PO once daily, clonidine 0.1 mg PO TID, metoprolol 50 mg PO BID, and amlodipine 10 mg PO once daily. On admission, the patient was initiated on aspirin 325 mg, atorvastatin 80 mg, and clopidogrel 75 mg PO daily. Her dose of clonidine was increased to 0.2 mg PO TID to optimize blood pressure control, and metoprolol and amlodipine were continued at the same doses. The patient developed fever on the third day after the cardiac catheterization. The fever ranged from 99.0°F to 102.7°F. The physical examination, laboratory data analysis, multiple blood cultures, urinalysis, chest radiograph, and a computed tomography of the head, chest, abdomen, and pelvis did not reveal any source of infection. On the sixth day after admission, clonidine was reduced to the baseline dose of 0.1 mg PO TID and on the ninth day it was stopped. The patient was afebrile on the twelfth day and remained so for the duration of her hospitalization. Naranjo scores for her newly initiated concomitant medications were as follows: aspirin, 1; atorvastatin, 3; clonidine, 6; and clopidogrel, 1. The rating of 6 for clonidine suggests that it was probably associated with the fever in this patient.

Conclusion:

We describe a case of drug-induced fever probably associated with clonidine administration. The higher dose of clonidine alone or in inter- action with duloxetine and atorvastatin may have contributed to the development of drug-induced fever.

INTRODUCTION

Drug-induced fever is a clinical diagnosis and should always be considered when the fever is constant and high without a clear source of infection.¹ Clonidine is a centrally acting antihypertensive agent, which decreases sympathetic outflow from the central nervous system to the peripheral tissues.^2–5 Although the final pathway is not completely understood, one of the mechanisms of sympathoinhibition is stimulation of α₂-adrenergic receptors and imidazoline receptors. Clonidine has been reported to have an effect on the regulation of temperature based on evidence from experimental studies^2–5 but has not been previously reported to be associated with fever in humans. We report a case of fever in a patient taking clonidine.

CASE SUMMARY

A 66-year-old white female nursing-home resident with a history of morbid obesity (body mass index [BMI], 40 kg/m²), Alzheimer’s dementia, hypertension, and depression presented with generalized weakness and shortness of breath to a hospital in Boston, Massachusetts (Caritas Saint Elizabeth’s Medical Center). She was found to have a non–ST-segment elevation myocardial infarction (diagnosed via elevated levels of troponin T [1.4 ng/mL] and absence of ischemic changes in ECG) and she underwent cardiac catheterization (anesthesia: fentanyl 100 μg IV and midazolam 6 mg IV) and percutaneous coronary intervention with placement of a stent in the left anterior descending artery. Prior to hospitalization, the patient had been taking memantine hydrochloride 10 mg PO BID and donepezil hydrochloride 10 mg PO once daily for dementia, duloxetine hydrochloride 60 mg PO once daily for depression, and clonidine 0.1 mg PO TID, metoprolol 50 mg PO BID, and amlodipine 10 mg PO once daily for hypertension for several years. Her blood pressure was inadequately controlled with the current regimen (systolic blood pressure range, 140–165 mm Hg; diastolic blood pressure range, 60–80 mm Hg, according to the records of the nursing home). Her heart rate ranged between 50 and 70 beats/min. She had no history of drug allergies. On admission, the patient was initiated on aspirin 325 mg, atorvastatin calcium 80 mg, and clopidogrel 75 mg PO once daily. Her dose of clonidine was increased to 0.2 mg PO TID to optimize blood pressure control. The patient developed fever the third day after cardiac catheterization (body temperature was assessed orally q4h). The fever was constant and ranged between 99°F and 102.7°F (Figure). The physical examination, laboratory data analysis, multiple blood cultures, urinalysis, and chest radiographs were unremarkable. Pertinent laboratory data included a white blood cell (WBC) count of 7200/μL with 70% polymorphonuclear cells, 25% lymphocytes, and 4% monocytes (Table). Her laboratory data did not have any clinically important change during her hospitalization, including her WBC count (Figure). An ECG (performed on admission) revealed sinus rhythm, left-axis deviation, and nonspecific ST-T wave changes. A computed tomography of the head, chest, abdomen, and pelvis (with intravenous contrast) was done on the sixth day of hospitalization and did not reveal any source of infection. Her fever was high and constant and, based on the pattern of the fever, the absence of any source of infection, and because the patient clinically remained hemodynamically stable with unchanged physical examination, a diagnosis of possible drug-induced fever was made and no antibiotics were initiated. Atorvastatin was discontinued on the fourth day after cardiac catheterization secondary to elevated creatine phosphokinase (CPK) (380 U/L). On the sixth day after admission, clonidine was reduced to the baseline dose of 0.1 mg PO TID. On the ninth day, clonidine was discontinued. The patient was afebrile on the twelfth day and remained afebrile for the duration of her hospitalization. She was discharged to her nursing home after 2 weeks of hospitalization.

Figure.

Temperature (oral) and white blood cell (WBC) count during hospitalization in a 166-year-old morbidly obese white female with Alzheimer’s dementia treated with clonidine, duloxetine, and atorvastatin who developed unexplained high fever. *Cardiac catheterization was performed and aspirin, clopidogrel, and atorvastatin were initiated, and clonidine dose was increased to 0.2 mg PO TID. ^†Atorvastatin was discontinued, secondary to elevated creatine phosphokinase and a computed tomography scan was conducted of the chest, abdomen, and pelvis. ^‡Clonidine was reduced to the baseline dose of 0.1 mg PO TID. ^§Clonidine was discontinued. ^∥Patient was afebrile and remained so for the duration to of her hospitalization.

Table.

Laboratory data (obtained on admission) from a 166-year-old morbidly obese white female with Alzheimer’s dementia treated with clonidine, duloxetine, and atorvastatin with unexplained high fever. Reference values of laboratory data were developed at the clinical laboratory of a hospital in Boston, Massachusetts (Caritas Saint Elizabeth’s Medical Center).

Laboratory Data	Recorded Value	Reference Value
Alanine aminotransferase	28 U/L	7–40 U/L
Albumin	3.4 g/dL	3.1–4.3 g/dL
Alkaline phosphatase	121 U/L	30–115 U/L
Aspartate aminotransaminase	45 U/L	5–35 U/L
Bicarbonate	33 mEq/L	21–34 mEq/L
Bilirubin, total	1.0 mg/dL	0–1.1 mg/dL
Blood urea nitrogen	17 mg/dL	8–25 mg/dL
Calcium	8.1 mg/dL	8.1–10.5 mg/dL
Chloride	102 mEq/L	96–112 mEq/L
Cholesterol, total	220 mg/dL	<240 mg/dL
Creatine phosphokinase	428 U/L	40–250 U/L (females)
Creatinine	0.7 mg/dL	0.6–1.2 mg/dL
Creatinine clearance	146 mL/min	80–145 mL/min
Ferritin	166 ng/mL	15–300 ng/mL
Folic acid	14.1 ng/dL	2.0–21 ng/dL
Glucose	113 mg/dL	65–100 mg/dL
HDL-C	34 mg/dL	30–70 mg/dL
Hemoglobin	10.6 g/dL	11–16 g/dL (females)
Iron	40 μg /dL	50–170 μg/dL
Lactic dehydrogenase	194 U/L	<260 U/L
LDL-C	148 mg/dL	<130 mg/dL
Magnesium	2.1 mg/dL	1.5–2.5 mg/dL
Platelets	426,000/μL	150,000–450,000/μL
Potassium	3.5 mEq/L	3.5–5.3 mEq/L
Sodium	142 mEq/L	135–145 mEq/L
Thyroid-stimulating hormone	4.30 μU/mL	<10 μU/mL
Total iron-binding capacity	221 μg/dL	250–450 μg/dL
Triglycerides	190 mg/dL	<140 mg/dL
Troponin T (peak)	1.4 ng/mL	0–0.2 ng/mL
Vitamin B12	452 pg/mL	200–800 pg/mL
White blood cell count	7200/μL	3800–11,000/μL

DISCUSSION

This case suggests that clonidine was a probable cause of drug-induced fever. Although drug-induced fever has been reported with other centrally acting antihypertensive drugs such as methyldopa, a centrally acting α₂-agonist,⁶ a search of the literature did not identify any previous reports of this adverse effect with clonidine in humans (MEDLINE [1980–August 2009], English-language literature, search terms fever, hyperthermia, temperature, clonidine, atorvastatin, duloxetine, and clopidogrel in different combinations). The references cited in the search results were examined to identify additional reports regarding the possible association of these drugs with development of fever.

The patient had been on memantine (an N-methyl-D-aspartate receptor antagonist), duloxetine (a serotonin-norepinephrine reuptake inhibitor), donepezil (a centrally acting reversible acetyl cholinesterase inhibitor), metoprolol (a β-blocker), and amlodipine (a long-acting calcium channel blocker) for many years and these agents have not been reported to be associated with development of fever. Memantine was reported to prevent the development of hyperthermia in an animal model for serotonin syndrome.⁷ Duloxetine was reported to cause eosinophilic pneumonia with fever in one human case.⁸ Thus, these drugs most likely were not associated with the new onset of fevers in this patient. The new medications that were initiated in our patient were aspirin, atorvastatin, clopidogrel, and increased dose of clonidine. Aspirin is known reduce fever and to our knowledge has not been associated with drug-induced fever. Statins have an anti-inflammatory effect and have been reported to reduce fever.⁹ Atorvastatin has been associated with a drug reaction associated with eosinophilia, systemic symptoms, and fever in one case.¹⁰ However, in our case fever was present for 1 week after the last dose of atorvastatin, there was no rash, and fever persisted longer than 4 times the t_1/2 of atorvastatin (~14 hours). Since a stent was recently placed in our patient, clopidogrel was continued during the entire hospitalization, even when the patient was afebrile and there was no association between the use of clopidogrel and development of fever. The patient developed fever 3 days after her cardiac catheterization and therefore, most likely, the anesthesia (patient received midazolam and fentanyl) or the cardiac catheterization was not associated with the development of fever. Of note, use of midazolam has not been reported to be associated with development of fever. Opioids, including fentanyl, have been reported to have an antipyretic effect in parturients.¹¹ Therefore, based on the development of fever 3 days after anesthesia, we ruled out the possibility of malignant hyperthermia. Although atypical neuroleptic malignant syndrome can present as fever of unknown origin,¹² the absence of rigidity, change in mental status, and normalization of CPK after discontinuation of atorvastatin, ruled out this possibility. The pattern of fevers did not seem to correlate with use of intravenous contrast.

Application of the Naranjo adverse drug reaction probability scale to this case (score of 6 for clonidine) suggested that the fever was probably associated with the higher dose of clonidine.¹³ Hyperthermia or high fever have not been previously reported with clonidine.¹⁴ Naranjo scores for her newly initiated concomitant medications were as follows: aspirin, 1; atorvastatin, 3; and clopidogrel, 1.

The patient had a normal WBC count during her hospitalization (<10,000/μL). There was an isolated count of 9900 on the thirteenth day, but the patient had a normal differential (75% polymorphonuclears, 23% lymphocytes). Therefore, we do not think that the slight increase of WBCs after discontinuation of clonidine was clinically significant in this case; the patient remained afebrile and all repeat cultures were negative.

The mechanism by which clonidine might disrupt thermoregulation is still unclear, but both central and peripheral mechanisms may be implicated. Clonidine, an α₂-adrenoceptor agonist, may induce hypothermia by action at central α_2A-adrenoceptors.^4,5 Animal studies using various α-adrenoceptor antagonists have suggested that the hypothermic response of clonidine is primarily mediated by activation of postsynaptic α_2A-adrenoceptors in the preoptic area of the hypothalamus.⁴ α_2A-Adrenoceptors are also found presynaptically as inhibitory receptors regulating release of noradrenaline and other neurotransmitters, such as dopamine and serotonin, in the central and peripheral nervous systems.² A study in rats found that clonidine-induced hyperthermia was mediated by postsynaptic α₂-adrenoceptors.¹⁵ Since serotonergic, noradrenergic, and dopaminergic neurotransmitter systems have all been implicated in the mediation of hypothermia and hyperthermia, acute increases in these neurotransmitters induced by medications may influence the thermoregulatory system. Concomitant activation of the presynaptic α_2A-adrenoceptor may result in a hyperthermic response.² In animal models, it has been found that the thermoregulatory effect of α₂-adrenoceptors can be affected by medications such as fluoxetine³ or streptozotocin.² Actions of other agents such as 3,4-methylenedioxymethamphetamine at α₂^-adrenoceptors may also alter the hypo/hyperthermic actions of clonidine.² In a diabetic model of rats, it was found that the thermoregulatory responses to clonidine can be altered with different and higher doses of clonidine and addition of streptozotocin, a naturally occurring antibiotic that is particularly toxic to the insulin-producing β-cells and can also affect α₂-receptor function, was associated with development of hyperthermia rather than hypothermia.¹⁶ Feleder et al¹⁷ reported in guinea pigs that the induction of postsynaptic prostaglandin E2 (PGE2) by presynaptic norepinephrine, mediated by α₂-adrenoreceptors and modulated by cyclooxygenase 2 (COX-2) and clonidine through stimulation of α₂-adrenoreceptors, can induce a late COX-2/PGE2-dependent increase in core temperature. Therefore, it is possible, in our case, that the interaction between the higher dose of clonidine and duloxetine might have led to increased core temperature and drug-induced fever. Another important contributing factor may have been that the patient was initiated on atorvastatin, an inhibitor of cytochrome P450, which might have resulted in less metabolism of duloxetine.^8,10 The increased concentration of duloxetine could have further facilitated the interaction between increased concentration of serotonin and clonidine. Moreover, it is unclear if obesity in our patient might have been a contributing factor that facilitated the drug-induced fever. Obesity is a proinflammatory state and increased BMI has been associated with higher levels of cytokines and a higher incidence of drug-induced fever in patients receiving clozapine.¹ Finally, centrally acting drugs may have different effects and interactions in sick patients, as was recently reported with clonidine and escitalopram.¹⁸

Clonidine should be considered in the differential diagnosis of drug-induced fever in the setting of polypharmacy and interactions with other medications including serotonin reuptake inhibitors and statins.

CONCLUSION

We describe a case of drug-induced fever probably related to clonidine administration. The higher dose of clonidine alone or in interaction with duloxetine and atorvastatin may have contributed to the development of drug-induced fever.